Maths Coursework: investigation into the correlation between IQ and KS2 results

Authors Avatar

Maths Coursework: investigation into the correlation between IQ and KS2 results


In this investigation I aim to investigate the correlation between IQ and the result obtained for KS2 result. This investigation will help to prove whether it is intelligence that affects the marks obtained in and examination, or if there is some other factor involved. If a conclusion is made, then education may be able to take a step further into improving the system, and helping children do better in examinations. I intend to work from a hypothesis, and by the end hopefully prove or disprove this hypothesis.


The hypothesis that I will be working from will be:

“The higher the IQ level of an individual, the higher mark / level, that individual will get in KS2 examinations.”



        To either prove or disprove this hypothesis, I have been given a data set of around 1200 students (1183 to be exact), from Mayfield School. I predict that the higher the IQ, the higher the individual’s score is likely to be, because their intelligence is higher, but that may also depend on another creative or artistic intelligence, for example, so this is

not entirely a firm prediction, but it what I expect to happen.


The way I will catty out the investigation will be to sort out the data into year groups, work out their percentages in comparison to the total, and use the figure as a number out of 100. Then I will further divide it into male and female, and do the same for each year group, so an equal representation, of not just the year groups, but male and females are also shown, to give an equal representation of the whole group of students.

The Statistics are as shown below:

*Out of a total of 1183 students      **Rounded to the nearest whole number

Above is the representation of the entire year groups, to improve this investigation further, I can include a further variable of race, which would also give a representation of races in the school, ,but the data for this is not provided, so it can not be tested.

After having sorted it into stratified sets of year group and male and female, I have randomly selected 100 samples, with the numbers as shown above. I have used the random key- ‘Ran#’, and after obtaining a number (to hundredths – three digits after the decimal point), I multiplied it by the total amount available. For instance in Year 7, there are 151 males. Let’s say for example I obtain 0.625 on the calculator random key. I then multiply it by 151 to get 94.375. This is then rounded up to the nearest whole number, because you can’t have half an entry. So I copy record number 94, and copy it into a separate sheet, and continue this for the entire data set.

Join now!

And after having done this, I have got the following set of data, which I will use to analyse, and prove or disprove my hypothesis, and analyse these using graphs. Firstly, I will analyse them individually to check for range and distribution, and then move on testing the hypothesis using scatter graphs. The final data set is as follows:

Data presentation / Analysis:

After having got the final data range, firstly, the standard deviation and correlation coefficient was calculated:

The standard deviation shows, on average, the distance of ...

This is a preview of the whole essay